Learning Weighting Map for Bit-Depth Expansion within a Rational Range
This work addresses the need for accurate and artifact-free image display in applications like digital media, though it is incremental as it builds on existing BDE methods with a novel optimization approach.
The paper tackles the problem of bit-depth expansion (BDE) for converting low bit-depth images to high bit-depth ones, proposing a bit restoration network (BRNet) that learns pixel-wise weighting maps to avoid altering high-order bits, resulting in improved performance with higher PSNR/SSIM and lower Wasserstein distance compared to state-of-the-art methods.
Bit-depth expansion (BDE) is one of the emerging technologies to display high bit-depth (HBD) image from low bit-depth (LBD) source. Existing BDE methods have no unified solution for various BDE situations, and directly learn a mapping for each pixel from LBD image to the desired value in HBD image, which may change the given high-order bits and lead to a huge deviation from the ground truth. In this paper, we design a bit restoration network (BRNet) to learn a weight for each pixel, which indicates the ratio of the replenished value within a rational range, invoking an accurate solution without modifying the given high-order bit information. To make the network adaptive for any bit-depth degradation, we investigate the issue in an optimization perspective and train the network under progressive training strategy for better performance. Moreover, we employ Wasserstein distance as a visual quality indicator to evaluate the difference of color distribution between restored image and the ground truth. Experimental results show our method can restore colorful images with fewer artifacts and false contours, and outperforms state-of-the-art methods with higher PSNR/SSIM results and lower Wasserstein distance. The source code will be made available at https://github.com/yuqing-liu-dut/bit-depth-expansion